Fact-checked by Grok 2 weeks ago

Lumped-element model

The lumped-element model (LEM) is a fundamental approximation in that represents complex physical systems, such as circuits and transmission lines, as networks of discrete, idealized components—including resistors, capacitors, inductors, and conductors—connected by perfect wires, under the assumption that the system's physical dimensions are much smaller than the of the operating signals. This approach simplifies the analysis of electromagnetic behavior by applying basic circuit laws, such as Kirchhoff's voltage and current laws, instead of solving full , making it suitable for low-frequency applications where signal propagation delays within elements are negligible. Central to the LEM are the key parameters that characterize each element: (R), representing energy dissipation; (C), storing electric energy; (L), storing ; and, in some cases, conductance (G) for leakage effects. For transmission lines, these are often expressed on a per-unit-length basis (R', G', C', L') to model distributed effects through infinitesimal segments, enabling the derivation of for wave propagation analysis. The model's validity relies on core assumptions, including quasi-static conditions where electric and magnetic fields are uniform within each element, the absence of significant radiative losses, and operation at frequencies low enough that the time for electromagnetic waves to traverse an element (approximately element length divided by the ) is much shorter than the signal period—typically holding for systems under a few hundred megahertz depending on scale. At higher frequencies or larger scales, the breaks down, necessitating distributed-parameter models that account for wave propagation along the structure. Beyond electrical circuits, the LEM extends to analogous domains like systems (masses as inductors, springs as capacitors, dashpots as resistors) and acoustics, facilitating simulations in fields such as microelectromechanical systems (), transducers, and even thermal energy transfer. Its advantages include computational simplicity and intuitive design, underpinning tools like SPICE simulators for predicting system behavior, though accuracy requires careful validation against physical measurements to ensure assumptions hold.

Fundamentals

Definition and principles

The lumped-element model is an approximation technique used to represent distributed physical systems, such as those involving continuous fields or media, by a collection of interconnected components known as "lumps" or elements. These elements are idealized as storing or dissipating without significant spatial variation within their boundaries, allowing systems to be simplified into manageable networks. This approach is particularly applicable when the system's dimensions are small compared to the characteristic wavelengths or scales of variation, enabling the neglect of propagation delays and field distributions. At its core, the lumped-element model describes through ordinary differential equations (ODEs) rather than partial differential equations (PDEs), as it assumes uniformity within each lump and treats time as the sole independent variable. The topology of the model is conceptualized as a comprising nodes (junctions where elements connect) and branches (the elements themselves), facilitating analysis via conservation laws analogous to those in . This framework unifies the behavior of diverse systems by mapping and dissipation to equivalent circuit-like structures, where variables such as voltage, , , or represent effort and flow quantities across domains. The model originated in 19th-century , building directly on formulated in 1845, which provided the foundational principles for analyzing networks of discrete components under the lumped approximation. These laws—governing the conservation of charge at nodes and energy around loops—were initially applied to electrical circuits but later extended to , , and other physical domains through analogy-based mappings, forming the basis of the lumped-matter discipline for multi-physics modeling. In mathematical terms, the model's equations take a general form reflecting balance in each , such as for a capacitive storage where flow i = C \frac{dv}{dt} (with C as , v as voltage, and t as time) or for inductive storage where effort v = L \frac{di}{dt} (with L as ). These are generalized across domains—for instance, to mass-spring s via F = m \frac{dv}{dt} for inertial elements—yielding a system of coupled ODEs solvable for the network's response.

Assumptions and validity conditions

The lumped-element model is predicated on the assumption that system components can be treated as point-like entities, where physical properties such as voltage, , or exhibit no significant internal spatial variation across the element. This idealization posits that each element behaves as a discrete , concentrating , dissipation, and generation at idealized points without distributed effects. Interactions between these elements are mediated solely through lumped state variables, such as voltage and in electrical systems or and in mechanical ones, enabling the representation of complex systems via interconnected ordinary differential equations. These assumptions hold under conditions where the characteristic size of the or its components is much smaller than the relevant of propagating phenomena, typically ensuring that the maximum dimension is less than one-tenth of the (λ/10) to neglect wave delays and variations. At low frequencies, where quasi-static approximations apply, the time for signals to traverse the is negligible compared to the of , justifying the absence of effects. This validity extends to scenarios where times are insignificant relative to the 's dynamic timescales, allowing assumptions across elements. The fails in distributed systems, such as high-frequency waveguides or antennas, where wave effects like and dominate due to dimensions comparable to or exceeding the , leading to inaccurate predictions of behavior. Similarly, in long lines, spatial variations in voltage and current arise from distributed and , invalidating the lumped paradigm. Even in compact devices, unmodeled parasitic elements—such as stray capacitances or inductances—can introduce substantial errors when operating near the limits of the . To assess applicability across domains, dimensionless criteria provide quantitative benchmarks for the feasibility of lumping. In thermal systems, the (Bi = hL/k, where h is the , L the , and k the thermal ) should satisfy Bi ≪ 0.1 to ensure negligible internal temperature gradients relative to surface , validating uniform temperature assumptions. For fluid systems, the (Kn = λ/L, where λ is the molecular and L the ) must be Kn < 10^{-3} to uphold the continuum hypothesis, beyond which rarefied gas effects necessitate non-lumped or kinetic descriptions. These metrics delineate the regime where the lumped-element approach reliably captures system dynamics without distributed corrections.

Lumped-matter discipline

The lumped-matter discipline serves as a unifying framework for applying lumped-element modeling to physical systems involving matter interactions, extending beyond electrical domains to encompass mechanical, thermal, fluid, and other multi-physics scenarios. This approach conceptualizes such systems as networks analogous to electrical circuits, where components are treated as discrete "lumps" interconnected to exchange energy. Tools like facilitate this multi-domain representation by diagramming energy flows and storage without domain-specific conventions, enabling a consistent analysis across disciplines. Central to the discipline is the application of universal conservation principles, akin to , to generalized variables known as efforts and flows. Efforts represent intensive quantities such as force in mechanical systems, pressure in fluids, or temperature differences in thermal contexts, while flows denote extensive rates like velocity, volume flow, or heat transfer rate. The product of effort and flow yields power, ensuring energy balance at junctions (flow conservation) and loops (effort summation), which underpin the modeling of interactions in matter-based systems. The discipline was formalized in the mid-20th century within system dynamics, notably by Henry Paynter at MIT, who introduced bond graphs in 1959 as a graphical language for deriving dynamic equations from physical structures. Paynter's work, detailed in his 1961 monograph, emphasized energy-based modeling to bridge domains, laying the groundwork for computational simulations. This development has influenced extensions of circuit simulation tools, such as adaptations of for multi-domain analysis, allowing engineers to model hybrid systems numerically. One key advantage of the lumped-matter discipline is its promotion of interdisciplinary integration, particularly in fields like , where mechanical actuators, electrical controls, and thermal effects must be analyzed cohesively. By abstracting diverse phenomena into a common topology, it simplifies the derivation of state-space models and control strategies, reducing errors in cross-domain simulations. The validity of these analogies relies on the assumptions outlined in the general principles of lumped modeling, ensuring negligible propagation delays within lumps.

Electrical systems

Basic elements and circuit topology

In the lumped-element model for electrical systems, the fundamental passive components are the resistor, capacitor, and inductor, each characterized by constitutive relations that relate voltage and current under the assumption of negligible spatial variations within the element. A resistor obeys Ohm's law, where the voltage drop v across it is proportional to the current i flowing through it, given by v = i R, with R denoting the resistance in ohms; this element dissipates energy as heat without storing it. A capacitor stores energy in an electric field and relates current to the rate of change of voltage via i = C \frac{dv}{dt}, where C is the capacitance in farads; ideally, it behaves as an open circuit at steady-state DC. An inductor stores energy in a magnetic field and follows v = L \frac{di}{dt}, with L as the inductance in henries; it acts as a short circuit at steady-state DC under ideal conditions. Ideal elements assume perfect behavior without parasitics, but real components exhibit non-ideal traits such as frequency-dependent resistance in inductors due to skin effect or leakage resistance in capacitors, which must be accounted for in high-frequency lumped models. Circuit topology in lumped models describes the interconnection of these elements through nodes (connection points), branches (element paths between nodes), and loops (closed paths of branches), enabling systematic analysis of network behavior. Kirchhoff's current law (KCL), formulated in 1845, states that the algebraic sum of currents entering a node equals zero, reflecting charge conservation. Kirchhoff's voltage law (KVL) asserts that the algebraic sum of voltages around any closed loop is zero, based on energy conservation in conservative fields. Lumped circuits incorporate both passive elements (resistors, capacitors, inductors, which cannot generate energy) and active elements (such as independent voltage or current sources that supply constant electromotive force or current regardless of the circuit, and dependent sources whose output depends on another voltage or current elsewhere in the network). Independent sources provide fixed values, like a battery modeled as an ideal voltage source, while dependent (or controlled) sources, such as voltage-controlled current sources, enable modeling of active devices like transistors in larger networks. Interconnections of these elements yield differential equations in the time domain for transient analysis; for instance, a series RC circuit driven by a step voltage input produces a first-order equation whose solution involves the time constant \tau = RC, representing the time for the capacitor voltage to reach approximately 63% of its final value. This approach, grounded in the lumped-matter discipline, facilitates analogies to other physical domains but is here applied strictly to electrical networks.

Modeling transmission lines and networks

Transmission lines, which are distributed-parameter systems, can be approximated using lumped-element models by representing them as an infinite ladder network consisting of repeating series and shunt elements. In this model, each infinitesimal section includes series resistance R and inductance L per unit length in the conductor path, along with shunt conductance G and capacitance C per unit length across the line. This structure captures the primary electrical effects of wave propagation in coaxial cables, microstrip lines, or twin-lead configurations, where R accounts for ohmic losses, L for magnetic energy storage, G for dielectric leakage, and C for electric energy storage. The lumped approximation divides the transmission line into small segments of length \Delta x, where each segment is modeled with discrete elements: series R \Delta x and L \Delta x, and shunt G \Delta x and C \Delta x. This discretization allows the application of standard circuit analysis techniques to approximate the behavior of the continuous line. The validity of this lumping requires that the segment length \Delta x and overall line length be much smaller than the signal wavelength \lambda (typically \Delta x \ll \lambda/10), ensuring that phase variations across each segment are negligible and the quasi-static assumption holds. Under these conditions, the model accurately predicts voltage and current distributions without significant wave retardation effects. The dynamics of this lumped model are governed by the telegrapher's equations in their differential form, derived from applied to the infinitesimal sections: \frac{\partial V(z,t)}{\partial z} = -(R + L \frac{\partial}{\partial t}) I(z,t) \frac{\partial I(z,t)}{\partial z} = -(G + C \frac{\partial}{\partial t}) V(z,t) These equations describe how voltage V and current I vary along the line position z and time t, incorporating both resistive and reactive effects. In the frequency domain, they simplify to \partial V / \partial z = -(R + j\omega L) I and \partial I / \partial z = -(G + j\omega C) V, enabling solutions for characteristic impedance Z_0 = \sqrt{(R + j\omega L)/(G + j\omega C)} and propagation constant. For more complex electrical networks, lumped-element models facilitate analysis of multi-port systems using impedance or state-space methods. Impedance methods represent the network via a Z-parameter matrix, where V_i = \sum Z_{ij} I_j relates port voltages to currents, allowing computation of transfer functions for interconnected components. State-space formulations, expressed as \dot{x} = A x + B u and y = C x + D u with state vector x capturing inductor currents and capacitor voltages, enable time-domain simulations of dynamic responses in software tools. Representative applications include low-pass filters, where cascaded LC sections approximate ideal responses, and impedance matching circuits, such as L-networks with series inductor and shunt capacitor to transform source impedance R_S to load R_L (e.g., let Q = \sqrt{\frac{R_L}{R_S} - 1}; then L = \frac{Q R_S}{\omega}, C = \frac{1}{\omega \frac{R_L}{Q}} for R_L > R_S). These techniques are widely used in RF design for amplifiers and antennas, prioritizing simplicity over full distributed analysis. At high frequencies, the lumped-element approximation breaks down when the line length approaches or exceeds \lambda/10, as phase shifts and retardation effects become prominent, leading to inaccuracies in predicting reflections and losses. In such cases, the model transitions to fully distributed representations, solving the as partial differential equations to account for wave propagation with finite velocity v = 1/\sqrt{LC}. This shift is essential for frequencies, where tools like or full-wave simulations replace lumped circuits to handle discontinuities and .

Mechanical systems

Analogies to electrical circuits

In mechanical systems modeled using the lumped-element approach, analogies to electrical circuits facilitate the of dynamic behavior through equivalent networks, leveraging established electrical techniques. These analogies map mechanical quantities and elements to their electrical counterparts, enabling the use of circuit theorems like Kirchhoff's laws for mechanical simulations. Grounded in the lumped-matter discipline, which treats components as discrete, idealized elements analogous to lumped electrical parts, these mappings preserve the mathematical structure of governing differential equations across domains. The , one of the earliest formal mappings, equates force to electrical voltage and velocity to electrical current. Under this correspondence, inertial mass acts like (storing ), viscous damping like (dissipating ), and compliance (1/k, where k is ) like (storing ). This analogy was first articulated in the context of acoustical and systems by Arthur G. Webster in , who drew parallels between and electrical impedances to describe wave propagation and system response.
Mechanical Quantity/ElementElectrical AnalogueRelation
Force (f)Voltage (V)f ↔ V
Velocity (v)Current (I)v ↔ I
Mass (m)Inductance (L)m ↔ L
Damper (b)Resistance (R)b ↔ R
Spring compliance (1/k)Capacitance (C)1/k ↔ C
An alternative formulation, known as the admittance or mobility analogy, reverses the primary variables by mapping force to current and velocity to voltage. In this scheme, mass corresponds to capacitance, damping to conductance (1/R), and spring stiffness to inductance, emphasizing mobility (velocity per unit force) as analogous to electrical admittance. Proposed by F. A. Firestone in 1933, this analogy addresses limitations in representing series-parallel mechanical connections and Kirchhoff's laws more intuitively for certain configurations, such as in acoustical transducers.
Mechanical Quantity/ElementElectrical AnalogueRelation
Force (f)Current (I)f ↔ I
Velocity (v)Voltage (V)v ↔ V
Mass (m)Capacitance (C)m ↔ C
Damper (b)Conductance (1/R)b ↔ 1/R
Spring stiffness (k)Inductance (L)k ↔ L
These analogies gained prominence in the 1940s, particularly for analyzing servomechanisms in military applications during World War II, where electrical network analyzers simulated mechanical control systems to predict stability and response. For instance, a classic mass-spring-damper system under the impedance analogy translates to a series L-C-R circuit, where the inductor represents mass, capacitor the spring, and resistor the damper; this equivalence allows mechanical transients to be solved using standard circuit simulation tools like SPICE precursors. Such representations, as detailed in early servomechanism studies, enabled engineers to apply electrical synthesis methods to mechanical design, improving feedback control in devices like gun turrets.

Applications in dynamics and vibrations

Lumped-element models are widely applied in the analysis of and , particularly for systems that can be approximated as multi-degree-of-freedom (MDOF) configurations treated as coupled oscillators. In such models, the system's masses, springs, and dampers represent inertial, , and dissipative elements, respectively, leading to a set of coupled equations that describe the dynamic behavior. The natural frequencies of these systems are determined by solving the eigenvalue problem derived from the undamped , where the eigenvalues correspond to the squares of the natural frequencies, providing insight into the system's modal properties and potential resonant conditions. Practical examples include vehicle suspension systems, often simplified as quarter-car models with lumped masses for the sprung and unsprung components, springs for and stiffness, and dampers for absorption, enabling the prediction of ride comfort and handling under road excitations. Similarly, in robotic arms, lumped-parameter approaches model flexible links as interconnected masses, springs, and dampers to capture vibration modes during motion, facilitating design for stability and precision. These models are frequently formulated in for control purposes, where position and velocity states are used to derive linear equations of the form \dot{x} = Ax + Bu and y = Cx + Du, allowing for feedback control strategies to suppress vibrations. Damping and resonance phenomena in these models are characterized by the quality factor Q, defined in mechanical terms as Q = \frac{\sqrt{km}}{b} for a single-degree-of-freedom system, where k is , m is , and b is coefficient, quantifying the sharpness of and energy dissipation per cycle. For forced , the steady-state response in the Laplace domain for velocity V(s) under force F(s) is given by V(s) = \frac{F(s)}{m s^2 + b s + k}, where b denotes the damping coefficient, highlighting amplification at resonance when the driving frequency approaches the natural frequency. Despite their utility, lumped-element models have limitations in mechanical dynamics, particularly for complex geometries or high-frequency responses where distributed effects like wave propagation become significant, necessitating the use of finite element methods (FEM) for more accurate of continuous structures. Lumped models are best suited for low-frequency regimes and simple geometries, while FEM excels in capturing spatial variations in stress and deformation across intricate shapes.

Thermal systems

Thermal resistance and capacitance

In thermal systems, the lumped-element model employs thermal resistance and capacitance as core components to represent heat conduction and storage, respectively, under the assumption of uniform within each lumped volume. Thermal resistance R_{th}, defined as R_{th} = \frac{\Delta T}{q} where \Delta T is the difference across the element and q is the steady-state flow rate, characterizes the 's opposition to conductive . For a plane wall undergoing one-dimensional conduction, this resistance takes the form R_{th} = \frac{L}{k A}, with L as the thickness, k as the thermal conductivity (a in W/m·K that quantifies the ease of conduction), and A as the cross-sectional area perpendicular to the flow. Thermal capacitance C_{th}, given by C_{th} = m c_p where m is the and c_p is the (a in J/kg·K representing the required to raise the of unit by one ), accounts for the storage of such that the stored is E = C_{th} \Delta T. The electrical analogy facilitates analysis by mapping temperature differences to voltages and heat flow rates to currents, enabling the use of familiar circuit concepts for thermal networks. In steady-state conduction, thermal circuits behave as resistive dividers, where series resistances add linearly (R_{th,total} = \sum R_{th,i}) and parallel paths combine reciprocally (\frac{1}{R_{th,total}} = \sum \frac{1}{R_{th,i}}), allowing heat flow to be computed as q = \frac{\Delta T_{overall}}{R_{th,total}}. This analogy extends to transient conditions through lumped thermal circuits, where physical objects are represented as capacitors interconnected by resistances; the capacitance at each node stores heat, while resistances govern conductive paths between nodes. For dynamic behavior, these RC-like models yield ordinary differential equations from energy balances at each node. The general form for a lumped node is C_{th} \frac{dT}{dt} = \sum q_{in} - \sum q_{out}, where incoming and outgoing heat flows are driven by temperature gradients across adjacent resistances (q = \frac{\Delta T}{R_{th}}). Solving such systems reveals time-dependent temperature profiles, often characterized by a time constant \tau = R_{th} C_{th}, analogous to electrical RC circuits and dependent on material properties k (influencing R_{th}) and c_p (scaling C_{th}). This framework aligns with the lumped-matter discipline, which provides a unified set of assumptions for modeling conserved quantities across domains like and charge.

Newton's law of cooling and heat transfer

states that the rate of convective q from a body to its surrounding ambient is proportional to the temperature difference between the body surface and the ambient, expressed as q = h A (T - T_\text{amb}), where h is the , A is the surface area, T is the body temperature, and T_\text{amb} is the ambient temperature. In lumped-element thermal models, this law is integrated to describe transient heat transfer under the assumption of uniform temperature within the body, valid when internal conduction resistance is negligible compared to surface convection resistance. The , defined as \text{Bi} = \frac{h L_c}{k}, quantifies this condition, where L_c is the (typically over surface area) and k is the ; the lumped approximation holds for \text{Bi} \ll 1, often \text{Bi} < 0.1, ensuring temperature uniformity. Applying an energy balance to the lumped body yields the ordinary differential equation \rho c V \frac{dT}{dt} = -h A (T - T_\text{amb}), where \rho is density, c is specific heat, and V is volume. The analytical solution for constant h and T_\text{amb} is an exponential decay: T(t) = T_\text{amb} + (T_0 - T_\text{amb}) e^{-t / \tau}, with time constant \tau = \frac{\rho c V}{h A}, representing the thermal response time of the system. For multi-element lumped systems, such as assemblies with internal heat generation or time-varying ambient conditions, the model extends to a set of coupled ordinary differential equations. Each lump's energy balance incorporates conduction between lumps (via thermal resistances), convection to ambient, and generation terms \dot{q}, resulting in a system solvable numerically or analytically for specific geometries, as in cylindrical heat-generating components.

Acoustics

Acoustic elements and impedance

In acoustic systems, lumped-element models approximate the behavior of sound waves under the assumption that wavelengths are much larger than the system's dimensions, allowing spatial variations to be neglected. This approach defines three fundamental elements: acoustic resistance R_a, which represents dissipative losses due to viscosity and thermal conduction; acoustic compliance C_a, which captures the compressibility of air volumes; and acoustic inertance M_a (also called acoustic mass), which accounts for the inertia of moving air masses. Acoustic resistance R_a is defined as the ratio of acoustic pressure difference \Delta p to volume velocity U (the product of particle velocity and cross-sectional area), R_a = \Delta p / U, with units of pascal-seconds per cubic meter (Pa·s/m³). It models energy dissipation in narrow passages, such as R_a = 8 \eta l / (\pi a^4) for a cylindrical tube, where \eta is air viscosity, l is length, and a is radius. Acoustic compliance C_a is the ratio of volume displacement to pressure difference, C_a = \Delta V / \Delta p, or equivalently in frequency domain, U = j \omega C_a \Delta p, with units of cubic meters per pascal (m³/Pa); for an enclosed air volume V, C_a = V / (\gamma P_0), where \gamma \approx 1.4 is the adiabatic index and P_0 is ambient pressure. Acoustic inertance M_a relates pressure to the acceleration of volume velocity, \Delta p = M_a \, d^2 U / dt^2, or in frequency domain \Delta p = j \omega M_a U, with units of kilograms per square meter (kg/m⁴); for a short tube, M_a = \rho_0 l / S, where \rho_0 is air density, l is effective length, and S is cross-sectional area. These elements form the basis of acoustic impedance Z_a = \Delta p / U, analogous to electrical impedance Z_e = V / I, where acoustic pressure corresponds to voltage and volume velocity to current. The impedance of resistance is Z_a = R_a (in phase with flow); for compliance, Z_a = 1 / (j \omega C_a) (pressure lags flow by 90°); and for inertance, Z_a = j \omega M_a (pressure leads flow by 90°). This analogy facilitates simulation using electrical circuit tools, with direct mapping: R_e = R_a, L_e = M_a, C_e = C_a. Acoustic impedance Z_a (in acoustic ohms, Pa·s/m³) differs from specific acoustic impedance z = p / u (in rayls, Pa·s/m²), where u is particle velocity; the former applies to lumped networks, while the latter describes plane-wave propagation in unbounded media. Basic acoustic networks combine these elements to model resonators and enclosures at low frequencies, where the lumped approximation holds (e.g., dimensions << wavelength). A classic example is the , consisting of a cavity (compliance C_a) connected via a neck (inertance M_a, often with series resistance R_a); its input impedance is Z_\text{in} = R_a + j \omega M_a + \frac{1}{j \omega C_a}, exhibiting resonance at \omega = 1 / \sqrt{M_a C_a} with minimum impedance R_a, analogous to a series . For closed enclosures without necks, the low-frequency model simplifies to a single compliance C_a = V / (\gamma P_0), representing uniform pressure buildup and volume oscillation, valid when the enclosure's longest dimension is much less than the wavelength. These networks enable analysis of sound absorption and transmission in devices like mufflers, using the for impedance scaling in simulations.

Modeling sound propagation and devices

Lumped-element models are applied to simulate sound propagation in acoustic systems where the dimensions are small compared to the wavelength, such as in ducts or enclosed rooms at low frequencies. In ducts, the model divides the structure into discrete segments, each represented by acoustic inertance for the air mass, compliance for compressibility, and resistance for viscous losses, allowing prediction of pressure and velocity along the path. For rooms or enclosures, the entire volume is often treated as a single compliant element coupled to radiating surfaces, capturing modal responses below the where modes are well-separated. To analyze complex systems like multi-segment ducts or lined enclosures, the transfer matrix method chains the impedance matrices of individual lumped elements, relating input pressure and volume velocity to outputs across the system. This approach efficiently computes transmission loss and resonance in perforated or resonator-lined ducts, as demonstrated in models of electromechanical acoustic liners. In practical devices, lumped models describe electroacoustic transducers like speakers and microphones. For dynamic speakers, the Thiele-Small parameters—such as resonance frequency f_s, total Q-factor Q_{ts}, and equivalent volume V_{as}—characterize the electromechanical coupling via a lumped circuit with voice coil resistance, suspension compliance, and cone mass, enabling enclosure design optimization. Enclosure effects, including back-cavity compliance and port inertance in bass-reflex systems, are incorporated to predict low-frequency response augmentation. Similarly, condenser microphones are modeled with diaphragm mass, acoustic compliance of the cavity, and capacitive transduction elements, quantifying sensitivity and directivity under small-signal conditions. Resonators like Helmholtz devices exemplify damping and resonance in these models, where the frequency is given by f = \frac{c}{2\pi} \sqrt{\frac{A}{V l}} with c the speed of sound, A the neck cross-sectional area, V the cavity volume, and l the effective neck length accounting for end corrections. Viscous and thermal losses introduce damping, broadening the resonance peak and reducing quality factor, critical for noise control applications. The lumped approximation holds when the enclosure or segment size is much less than the wavelength (\lambda / 10 or smaller), but transitions to distributed models are required near frequencies where wavelength approaches the physical dimensions, necessitating wave equations to capture higher-order modes and reflections.

Fluid systems

Fluid flow analogies

In lumped-element modeling of fluid systems, electrical analogies provide a powerful framework for analyzing flow dynamics by mapping hydraulic variables to their electrical counterparts. Pressure difference (ΔP) is analogous to voltage (V), while volume flow rate (Q) corresponds to electric current (I). This analogy allows fluid networks to be represented as electrical circuits, where fluid elements behave like resistors, inductors, and capacitors, facilitating the use of circuit analysis techniques for hydraulic and pneumatic systems. The basic fluid elements are defined through these analogies. Hydraulic resistance R_h, akin to electrical resistance, quantifies the pressure drop per unit flow rate and is given by R_h = \frac{\Delta P}{Q}, often derived from viscous effects in elements like orifices or narrow pipes. Inertance I_h, the fluid analog of inductance, arises from the inertial mass of the fluid and is expressed as I_h = \frac{\rho l}{A}, where \rho is fluid density, l is length, and A is cross-sectional area; it relates pressure drop to the rate of change of flow, \Delta P = I_h \frac{dQ}{dt}. Compliance C_h, corresponding to capacitance, measures the stored volume per unit pressure change due to fluid compressibility and is formulated as C_h = \frac{V}{\beta}, with V as volume and \beta as bulk modulus, capturing expansion in compliant components like accumulators. Fluid networks differ based on the working medium: hydraulic systems typically model incompressible liquids, where compliance is minimal and neglected, treating pipes as resistors or inductors for steady or transient flows; pneumatic systems, using compressible gases, emphasize compliance effects, making capacitance prominent in modeling expansion and compression. Orifices act primarily as resistors due to frictional losses, while long pipes contribute both resistance and inertance. These analogies extend within the lumped-matter discipline, which unifies multi-physics modeling by assuming uniform properties within elements across domains. The governing equations mirror Kirchhoff's laws: the sum of pressure drops around a closed loop equals zero (analogous to Kirchhoff's voltage law), ensuring energy conservation in fluid paths; at junctions, the algebraic sum of inflows equals outflows (analogous to Kirchhoff's current law), preserving mass. These relations enable solving lumped fluid circuits via nodal or mesh analysis. Multi-domain extensions integrate fluid-electrical interfaces, such as in electro-hydraulic actuators, where electrical signals control solenoid valves that modulate fluid flow, modeled by coupling circuit equations with hydraulic inertance and resistance at the interface. Bond graph methods formalize these hybrid systems, linking effort (pressure/voltage) and flow variables across domains.

Pipeline and hydraulic network modeling

In lumped-element modeling of pipelines, long fluid conduits are approximated by dividing them into discrete segments, each represented by concentrated parameters that capture key physical effects such as friction, inertia, and compressibility. Friction losses within each segment are typically modeled using the Darcy-Weisbach equation, which quantifies the pressure drop ΔP as ΔP = f (L/D) (ρ v² / 2), where f is the friction factor, L is the segment length, D is the pipe diameter, ρ is the fluid density, and v is the flow velocity. This approach allows for ordinary differential equations to describe segment dynamics, with resistance elements incorporating the nonlinear dependence of f on the Reynolds number Re = ρ v D / μ (μ being dynamic viscosity) to distinguish laminar (Re < 2300) from turbulent (Re > 4000) regimes. Surge effects in pipelines arise from , modeled via inertance elements analogous to inductors in electrical circuits, with the inertance value given by L = ρ L_s / A, where L_s is the segment length and A is the cross-sectional area. This inertance captures the required to accelerate or decelerate the , leading to oscillations during transients. In hydraulic networks, such as systems or hydraulic circuits, pipelines form interconnected graphs where nodes represent junctions (e.g., valves or pumps) and edges denote pipe segments with their lumped parameters. pressures and flows are solved using graph-based methods, such as matrices in the Laplace domain, enabling efficient simulation of steady-state and slow transient behaviors through matrix inversion or iterative solvers. For instance, in networks like those in systems, these models predict flow under varying demands, while in , they optimize response times. Transient phenomena, such as from sudden valve closure, are represented as oscillatory interactions between inertance (L) and capacitance (C = V / (ρ c²), where V is and c is wave speed) elements, akin to an with √(1/(L C)). This yields damped pressure waves propagating at speed c, with peak pressures up to ρ c Δv (Δv being change). Surge tanks mitigate these effects by acting as additional capacitors, storing to absorb oscillations and stabilize pressures in hydroelectric or long transmission lines. The validity of lumped models requires that segment lengths be much smaller than the shortest of interest (L_s << c / f_max, where f_max is the highest ), ensuring spatial uniformity within segments, and applies best to low-frequency transients where pipe length is short compared to wave travel time scales. Additionally, the determines the friction model's accuracy, with turbulent flows (high Re) necessitating empirical f correlations like Colebrook-White for reliable predictions.

Building heat transfer

Lumped models for envelopes and zones

In building heat transfer analysis, the envelope—comprising walls, roofs, and floors—is often modeled using lumped-element RC (resistance-capacitance) networks to simplify transient conduction processes. These models represent multi-layered constructions as series and parallel combinations of thermal resistances for conduction through materials and capacitances for storing heat in mass elements, such as surface layers exposed to . For instance, a common 3R2C configuration divides the wall into outer resistance (to external ), inner resistance (to internal ), a central conduction resistance, and two capacitances for the outer and inner surfaces. This approach approximates the distributed thermal behavior under the condition that the (Bi = hL/k, where h is the convective , L is the , and k is thermal conductivity) is much less than 1 (typically Bi < 0.1), ensuring uniform temperature within lumps. The overall heat transfer performance of these envelope models is quantified via the U-value, defined as the inverse of the total thermal resistance (U = 1/ΣR), which aggregates conduction resistances and surface film coefficients to predict steady-state heat loss. In practice, such RC networks facilitate dynamic simulations by solving ordinary differential equations for node temperatures, enabling predictions of envelope response to varying outdoor conditions without full finite-element analysis. For interior spaces, lumped models treat rooms as single-zone capacitors, where the air volume acts as a lumped with capacitance C = ρV c_p (ρ , V volume, c_p specific ), coupled to the envelope and accounting for heat inflows. Infiltration, modeled as an additional convective or airflow rate term, introduces exogenous air gains or losses, while gains are incorporated as time-varying sources applied directly to the zone node, often derived from and data. These single-zone formulations, as standardized in ISO 52016-1 (successor to ISO 13790), use simplified 5R1C networks to balance computational efficiency with accuracy for energy performance calculations. Extending to multi-zone buildings, networks connect individual zone capacitors via inter-zone resistances for conductive through partitions and elements for advective . Ventilation is represented as lumped fluid flow paths with associated resistances, linking zones to outdoor or supply air nodes, allowing of pressure-driven infiltration and distribution effects. This graph-based approach, where zones are nodes and couplings are edges, supports whole-building while maintaining lumped assumptions for each space. Such models are integral to energy standards like , which reference simplified thermal networks for compliance s, ensuring lumped approximations apply where Bi << 1 for envelope components to validate efficacy and interactions.

Energy balance and simulation applications

In lumped-element models for building s, the energy balance equation governs the temporal variation of within the air volume, expressed as \frac{dQ}{dt} = \sum q_{\text{in}} - \sum q_{\text{out}}, where Q represents the stored , q_{\text{in}} includes inflows such as gains, internal from occupants and , and controlled HVAC supply air, while q_{\text{out}} encompasses losses via conduction through surfaces, infiltration, and exhaust. This formulation treats the zone air as a well-mixed, lumped , simplifying spatial temperature variations while integrating component models from envelopes and zones as boundary conditions for fluxes. HVAC systems act as controlled elements in this balance, providing dynamic heating or cooling inputs based on setpoints and system performance curves to maintain desired conditions. Simulation of these models typically involves of the resulting equations (ODEs) over discrete time steps, often ranging from 1 to 15 minutes, to predict transient thermal behavior across design or operational periods. Tools like EnergyPlus employ lumped-zone representations within a modular heat balance framework, solving the zone air energy equation using methods such as third-order backward differencing or analytical solutions for linear systems, which ensure stability and accuracy in coupling with HVAC and balances. This approach enables whole-building simulations that account for interactions between zones, systems, and data, facilitating rapid iteration for . Applications of these energy balance simulations extend to assessing through indices like the Predicted Mean Vote (PMV), which quantifies occupant sensation on a -3 to +3 scale based on zone air , mean radiant , humidity, and activity levels derived from the lumped model outputs. They are also essential for heating and cooling load calculations, determining peak demands and system sizing to meet comfort criteria under varying conditions. However, lumping assumptions introduce uncertainties, such as overestimation of uniformity in stratified flows or simplified surface interactions, which can propagate errors in annual energy predictions depending on building geometry and . For more detailed facades where thermal gradients are significant, advanced approaches combine lumped-zone balances with distributed-parameter models, such as finite-difference methods for envelope layers, to capture non-uniform while maintaining computational efficiency for system-level simulations. This integration improves accuracy in scenarios like high-performance glazing or ventilated facades, reducing discrepancies between lumped predictions and measured data by incorporating CFD-derived distributions as inputs to the zone balance.

References

  1. [1]
    [PDF] 6.200 Lecture Notes: Lumped-Element Abstraction
    Feb 7, 2023 · Glossary. Lumped-Element Abstraction/Approximation In this system, lumped el- ements are used to model complex physical systems. The behavior ...
  2. [2]
    3.4: Lumped-Element Model - Engineering LibreTexts
    Sep 12, 2022 · It is possible to ascertain the relevant behaviors of a transmission line using elementary circuit theory applied to a differential-length lumped-element model.
  3. [3]
    Lumped Models - Stanford CCRMA
    The term ``lumped'' comes from electrical engineering, and refers to lumped-parameter analysis, as opposed to distributed-parameter analysis. Examples of ...
  4. [4]
    Lumped Element Model | Ultimate Electronics Book
    The lumped element model is a set of assumptions that let us treat various systems as being a single element that behaves like a single entity.
  5. [5]
    Lumped Parameter Modelling with Ordinary Differential Equations
    Mar 10, 2017 · This chapter presents an overview of ordinary differential equations and their use in lumped parameter modelling of physical systems and ...
  6. [6]
    Kirchhoff's Current and Voltage Laws | UniSA Online
    ... lumped element model of electrical circuits. They were first described in 1845 by German physicist Gustav Kirchhoff which generalized the work of Georg Ohm ...
  7. [7]
    Microwaves101 | Lumped Elements - Microwave Encyclopedia
    Lumped Elements. This page serves as the main page for our discussion on capacitors, resistors and inductors at microwave frequencies.
  8. [8]
    Lumped Circuit Analysis Vs Distributed Analysis - Rahsoft
    Jan 18, 2023 · The lumped element model is valid whenever Lc (the dimension of the component) is much less than the wavelength. Mostly the assumption is ...
  9. [9]
    [PDF] Electrical Design Worst-Case Circuit Analysis: Guidelines and Draft ...
    Jun 3, 2013 · When the dimensions of the physical infrastructure are on the order of the circuit's operational frequency wavelength, the standard lumped ...
  10. [10]
    [PDF] Full-wave Electromagnetic Modeling of Electronic Device Parasitics ...
    ... wavelength ratio becomes larger than 10% and the validity of those commonly-used lumped-element models becomes progressively more questionable. Instead, the ...
  11. [11]
    [PDF] Graphical Tools for Modelling Physical Systems - DSpace@MIT
    Aug 10, 1995 · The bond graph representation was developed at M.I.T. in 1959 by Henry M. Paynter in the Mechanical Engineering Department. Bond graph ...
  12. [12]
    Hierarchical semantic composition of biosimulation models using ...
    Bond graphs were invented by Henry Paynter and were initially meant to be used in mechanical systems [24]. In bond graphs, all systems are reduced to a ...
  13. [13]
    [PDF] Lumped Circuit Analysis
    For a resistor, V = IR, for an inductor V = L dI/dt and for a capacitor, I = C dV/dt. When the physical size of the circuit plays a role we're in the realm of.
  14. [14]
    [PDF] Module 5: Non-Ideal Behavior of Circuit Components
    Most engineers are introduced to common circuit elements such as resistors, capacitors, and inductors through a course in basic circuit analysis.
  15. [15]
    (PDF) Kirchhoff's Laws: Current and Voltage Rules in Electric Circuits
    Oct 2, 2025 · Formulated by Gustav Kirchhoff in 1845, these laws extend the principles of Ohm's Law to complex circuits, providing essential tools for ...
  16. [16]
    Voltage and Current Sources (Independent and Dependent Sources)
    Jul 17, 2020 · First, in general, power sources, both voltage and current sources, are active elements. This means they can generate energy as well as absorb, ...
  17. [17]
    [PDF] RC Circuits - Oregon State University
    Time-Domain Analysis of the RC Circuit. The behavior of this "RC" circuit can be analyzed in the time-domain by solving an appropriate differential equation ...
  18. [18]
    [PDF] 2.1 -The Lumped Element Circuit Model for Transmission Lines
    Jan 20, 2005 · The lumped element circuit model for transmission lines includes inductance, resistance, capacitance, and conductance, which are resistance, ...
  19. [19]
    [PDF] Chapter 7: TEM Transmission Lines - MIT OpenCourseWare
    Apr 7, 2022 · In this case the approximate Telegrapher's equations. (8.3.3–4) are often used. Example 7.1B. What is the characteristic impedance Zo for the ...
  20. [20]
    None
    ### Summary of High-Frequency Considerations for Lumped Models in Transmission Lines (Purdue ECE Lecture)
  21. [21]
    [PDF] MICROWAVE AND RF DESIGN NETWORKS - NC State Repository
    Matching using lumped elements is based on the impedance and admittance ... Figure 6-45: Matching network topologies using lumped elements and microstrip lines.
  22. [22]
    [PDF] Novel Methods and Algorithms for Fitting and Equivalent Circuit ...
    or reduced state-space representation, multi-port network synthesis is performed, as covered in the Chapter 4. The results and discussion are in Chapters 5 and ...
  23. [23]
    On the origins and development of mobility and impedance methods ...
    An account of the conception and early development of mobility and impedance methods in structural dynamics is presented in this paper.
  24. [24]
    A NEW ANALOGY BETWEEN MECHANICAL AND ELECTRICAL ...
    The new analogy is derived from the similarity of the following equations: v = fz̄ and E = IZ where z̄ is the reciprocal of the mechanical impedance as usually ...
  25. [25]
    [PDF] Unit 22 Vibration of Multi Degree-Of- Freedom Systems
    k − ω2 m = 0. (22-5). ~. ~. There will be n eigenvalues for an n degree-of-freedom system. In this case: eigenvalue = natural frequency.
  26. [26]
    [PDF] Structural Dynamics of Linear Elastic Multiple-Degrees-of-Freedom ...
    Through the solution of an eigenvalue problem, the free vibration shapes and their natural frequencies are obtained. Note that there will be n independent ...
  27. [27]
    [PDF] Vibration Analysis Projects of Lumped-Parameter and Distributed ...
    This paper documents an effort of integrating two FEA-based projects in the mechanical vibration course. The first project makes use of quarter-car models to.
  28. [28]
    Lumped Model of a Vehicle Suspension System - COMSOL
    A transient analysis is performed to compute the vehicle motion as well as the seat vibration levels for a given road profile.
  29. [29]
    Modeling, Controllability and Vibration Suppression of 3D Flexible ...
    Modeling, controllability and vibration suppression control of the 3D flexible robot are exclusively discussed in this paper. A lumped-parameter technique ...
  30. [30]
    Comparison of finite element and lumped parameter methods for ...
    The purpose of this work is to examine the basic foundations, similarities and fundamental differences between the finite element methods (FEM) and the ...
  31. [31]
    How to Use Lumped Elements to Model a Mechanical System
    May 29, 2019 · In general, all of the components of a mechanical system can essentially be modeled as a combination of lumped mass, stiffness, and damping ...
  32. [32]
    None
    Below is a comprehensive merged summary of "Thermal Resistance, Capacitance, and Electrical Analogy" from *Introduction to Heat Transfer, 6th Edition* by Bergman and Incropera. To retain all the detailed information from the provided segments efficiently, I will use a structured format with tables where appropriate, followed by a narrative summary. This ensures all page references, formulas, material properties, and key concepts are preserved while avoiding redundancy.
  33. [33]
    None
    ### Summary of Thermal Resistance and Capacitance in Lumped Models from Chapter 6
  34. [34]
    [PDF] fundamentals-of-heat-and-mass-transfer-incropera.pdf
    be a solid or a fluid, we use the term conduction to refer to the heat transfer that will occur across the medium. In contrast, the term convection refers ...
  35. [35]
    (PDF) Lumped Parameter Thermal Model of Heat-Generating ...
    In this study, an equivalent thermal network for a general geometry with internal heating in cylindrical coordinates is derived. The equivalent thermal circuit ...
  36. [36]
    [PDF] Lecture 7: Lumped Elements I. What is a lumped element? Lumped ...
    The region where the tube can act as a lumped element is the region where the pressure amplitude is nearly constant and the 'volume velocity' (v x tube cross-.
  37. [37]
    Acoustic compliance, inertance and impedance - Physclips.
    The compressibility of a small volume gives it an acoustic compliance; its inertia gives it an acoustic inertance. The ratio of acoustic pressure to flow is ...
  38. [38]
    Acoustic impedance and intensity: From Physclips Waves and Sound
    acoustic impedance = pressure/flow and specific acoustic impedance = pressure/velocity. The specific acoustic impedance, z, is an intensive property of a medium ...
  39. [39]
    [PDF] Acoustic Waveguides - University of Colorado Boulder
    (e) Inertance circuit element. (f) Compliance circuit element. (g) Lumped element model for a waveguide section. What is the equation of motion ...
  40. [40]
    Acoustic Property - an overview | ScienceDirect Topics
    A fluid region may be considered to be small, and treated as a lumped acoustic element, if all its principal dimensions are very much less than an acoustic ...
  41. [41]
    [PDF] Development of a Tunable Electromechanical Acoustic Liner for ...
    4 Lumped Element Model and Transfer Matrix of the EMHR ... Illustration of the modeling EMHR using the transfer matrix method, where each subsystem is ...
  42. [42]
    [PDF] 20100017134.pdf - NASA Technical Reports Server
    TRANSFER MATRIX FOR A DUCT ELEMENT. The transfer matrix for a straight duct was derived for a wave propagating through the duct with and without a mean flow ...
  43. [43]
    [PDF] Measurement of Thiele Small parameters at Scan-Speak
    Scan-Speak measures Thiele/Small parameters by measuring small signal parameters, including parameters like DC resistance, resonance frequency, and moving mass ...
  44. [44]
    [PDF] Lumped Parameter Model of a Loudspeaker
    Dec 14, 2020 · [14]. This measurement technique will yield the Thiele/Small parameters which can be used in a LPM of the loudspeaker.
  45. [45]
    Lumped-parameters equivalent circuit for condenser microphones ...
    Oct 17, 2017 · This work presents a lumped parameters equivalent model of condenser microphone based on analogies between acoustic, mechanical, fluidic, ...Missing: transducers | Show results with:transducers
  46. [46]
    Helmholtz Resonance - UNSW
    Later, we derive the equation for the frequency of the Helmholtz resonance. The vibration here is due to the 'springiness' of air: when you compress a ...
  47. [47]
    [PDF] Resonance Frequencies and Sound Absorption of Helmholtz ...
    Oct 5, 2018 · The proposed theoretical model for calculating the resonance frequency of a Helmholtz resonator with multiple necks via Eq. (9) is validated ...
  48. [48]
    Acoustic transmission line based modelling of microscaled channels ...
    Feb 21, 2019 · These dimensions are close to the wavelengths in the low ultrasonic range. Hence, distributed models like TLs have to be used instead of lumped ...
  49. [49]
    Analysis of lumped parameter models for blood flow simulations and ...
    Mar 24, 2004 · These lumped models can be derived from electrical circuit analogies where current represents arterial blood flow-rate and voltage represents ...
  50. [50]
    Hydraulic Resistance Calculator
    Jun 5, 2024 · Hydraulic Resistance Formula · R is the hydraulic resistance (Pa·s/m³) · ΔP is the pressure drop (Pa) · Q is the flow rate (m³/s).
  51. [51]
    [PDF] Fluid Power System Dynamics | Oylair
    Fluid power is the transmission of forces and motions using a confined, pressurized fluid, ideal for high speed, high force, and high power applications.
  52. [52]
    8. Bond Graph Models for Hydraulic Systems
    The generalized BG elements and relations apply to the modelling of dynamics of hydraulic systems in a similar way that the mechanical or electrical systems ...
  53. [53]
    Dynamic characteristics and sensitivity of water hammer problems in ...
    Jul 2, 2025 · This paper aims to use the Lattice Boltzmann Method (LBM) with entropy limiter to study the water hammer problems in propellant feedlines.
  54. [54]
    (PDF) The History of the Darcy-Weisbach Equation for Pipe Flow ...
    A concise examination of the evolution of the equation itself and the Darcy friction factor is presented from their inception to the present day.
  55. [55]
    [PDF] Modeling and Simulation of Hydraulic Long Transmission Line by ...
    Same elements for modeling of the lumped elements were used in both models. Using capacitor (C element), we model fluid compressibility, using coil (L element).
  56. [56]
    [PDF] Modelling and control of pipeline networks supplied by automated ...
    A model-based approach to control system design is developed for regulating the discharge flows at the outlets of a pipeline network that is supplied by an ...
  57. [57]
    Lumped Inertia Pipe with Orifice Waterhammer Theory
    Sep 24, 2025 · Such an element may also include an orifice as shown in Figure 1. These connector pipes are modeled as pipes with lumped inertia (Wylie, 1993.
  58. [58]
    Lumped parameter models for building thermal modelling
    A new, more accurate, analytic method is presented for creating the parameters of a second order LPM, consisting of three resistors and two capacitors.
  59. [59]
    Simplified Building Thermal Model Development and Parameters ...
    This paper proposes an approach to develop building dynamic thermal models that are of paramount importance for controller application.
  60. [60]
    Development of a Space Heating Model Suitable for the Automated ...
    Feb 2, 2019 · The ISO 52016-1:2017 [10] standard employs an RC network to calculate hourly internal air temperatures, hourly heating and cooling loads, and ...
  61. [61]
    Methodology for Modeling Multiple Non-Homogeneous Thermal ...
    Mar 14, 2023 · This paper presents a methodology for modeling structures with multiple thermal zones using the graph theory arrangement.
  62. [62]
    Inside Heat Balance: Engineering Reference — EnergyPlus 9.2
    This heat balance is generally modeled with four coupled heat transfer components: 1) conduction through the building element, 2) convection to the air, 3) ...
  63. [63]
    [PDF] Engineering Reference - EnergyPlus
    Mar 29, 2022 · ... heat balance is done at the Zone level. • Internal gains are modeled at the Space level and combined for the Zone heat balance. • Each Space ...
  64. [64]
    [PDF] Engineering Reference — EnergyPlus 8.0 - ETRM
    Each simulation time step, EnergyPlus performs a zone air heat balance to determine if cooling or heating is required to meet the zone thermostat set points,.
  65. [65]
    Hybrid Model: Engineering Reference — EnergyPlus 8.9
    EnergyPlus provides heat balance solution algorithms of 3rd order backward difference and analytical solution to solve the zone air energy balance equation.
  66. [66]
    Occupant Thermal Comfort: Engineering Reference — EnergyPlus 9.5
    Thermal comfort modeling in EnergyPlus is controlled primarily by the People input object. This includes input for selecting the type of thermal comfort model ...
  67. [67]
    [PDF] November 12, 2019 Building energy balances
    Nov 12, 2019 · What are energy balances useful for? 1. Heating and cooling load calculations. – i.e., equipment sizing to meet peak heating/cooling loads.
  68. [68]
    Building Energy Prediction Models and Related Uncertainties - MDPI
    This study provides a comprehensive review of building energy prediction models and uncertainties in the models.
  69. [69]
    [PDF] coupled energyplus and computational fluid dynamics natural ...
    In addressing the issue, a coupled lumped heat transfer model (EnergyPlus) and CFD model (Flu- ent) was implemented, and 8 days of simulation was con- ducted.
  70. [70]
    [PDF] Empirical Validation of Building Simulation Software - IEA EBC
    This report is a product of a joint effort between International Energy Agency Solar Heating and Cooling (IEA. SHC) Task 34 and Energy Conservation in Buildings ...